IBM PC/5150 Memory Issues

[ButINeededThatName]

I recently acquired a "presumed dead" 5150 (64kb-256kb system board) for free earlier this week. Turns out the generic-brand high-wattage PSU someone swapped into it had died, (what an absolute surprise) and swapping in an IBM branded unit brought it right back to life. Since then though, I've been chasing memory errors. At this point, I'm fairly certain one of the soldered on RAM chips have given up on life, though I wanted to get a second opinion before I shelve this system (I don't have any of the right tools to safely de-solder ICs).

Currently, I have pulled all of the RAM chips out of Banks 1/2/ & 3, leaving the soldered-on 64k at Bank 0 as the only installed system memory. I have switches 3 & 4 on block 1 both set to the ON position (Bank 0 populated only) and switches 1/2/3/4/ & 5 on block 2 all set to the ON position (64kb RAM installed). Upon POST, I receive errors 1055 201 (Bit Error and Memory Test Failed) and weirdly, 131 (Cassette Wrap Test Failed). If I try and boot to DOS from drive A, I receive "Parity Check 1" and the system locks up, requiring a hard-reset. If I remove the DOS boot disk the system boots to BASIC but only shows 12252 bytes of memory available.

Again, I'm pretty sure this is indicative of a bad memory IC on Bank 0, but if there's something else I can try to resolve this, I'm open to suggestions.

 

[modem7]

Currently, I have pulled all of the RAM chips out of Banks 1/2/ & 3, leaving the soldered-on 64k at Bank 0 as the only installed system memory. I have switches 3 & 4 on block 1 both set to the ON position (Bank 0 populated only) and switches 1/2/3/4/ & 5 on block 2 all set to the ON position (64kb RAM installed). Upon POST, I receive errors 1055 201 (Bit Error and Memory Test Failed) ...

For what you have just done on a 5150 motherboard of type 64KB-256KB, the 1055 201 is the expected result!
Expected, because of bugs in the 10/27/82 revision of motherboard BIOS.
For the 10/27/82 BIOS, you need to populate all motherboard banks, and set the motherboard switches accordingly.

... and weirdly, 131 (Cassette Wrap Test Failed).

Per [here], one known cause is a lack of minus 5 volts from the PSU.

If I try and boot to DOS from drive A, I receive "Parity Check 1" and the system locks up, requiring a hard-reset.

Right now, in the present configuration of your motherboard, the aforementioned bugs make DOS believe that only 16 KB is present.
With a belief that only 16 KB is present, the situation at [here] comes into effect.

If I remove the DOS boot disk the system boots to BASIC but only shows 12252 bytes of memory available.

The aforementioned bugs make Cassette BASIC believe that only 16 KB is present. Of that 16 KB, Cassette BASIC has reserved some for itself, leaving 12252 bytes.

 

[ButINeededThatName]

For what you have just done on a 5150 motherboard of type 64KB-256KB, the 1055 201 is the expected result!
Expected, because of bugs in the 10/27/82 revision of motherboard BIOS.
For the 10/27/82 BIOS, you need to populate all motherboard banks, and set the motherboard switches accordingly.

A strange bug to have, that's for sure. Anyways, I populated each bank one at a time and let the system boot to BASIC and each time I got the "correct" amount (28.5k for two banks, 45k for three) and finally, after populating the fourth, it worked! Prior to this I was getting a 4098 201 error, despite everything being set correct for 256k. Guess the chips needed a re-seat.

Per [here], one known cause is a lack of minus 5 volts from the PSU.

-5V pin on the connector reads -5.1, so no issues there. Testing pins 1 & 3 on the cassette port gives a reading of zero Ohms, which according the site means the motor relay is stuck in the "on" position. Not sure how I'd go about fixing that quite yet, though.

 

[modem7]

... and weirdly, 131 (Cassette Wrap Test Failed).

Per [here], one known cause is a lack of minus 5 volts from the PSU.

-5V pin on the connector reads -5.1, so no issues there. Testing pins 1 & 3 on the cassette port gives a reading of zero Ohms, which according the site means the motor relay is stuck in the "on" position. Not sure how I'd go about fixing that quite yet, though.

The motherboard might have a fault that results in the relay being permanently turned on. To see if that's the case, you will need to do some measurements. Refer to the diagram at [here], which shows what is expected to be observed when the motherboard is simply powered on. Are you able to make those measurements?

 

[modem7]

Testing pins 1 & 3 on the cassette port gives a reading of zero Ohms, which according the site means the motor relay is stuck in the "on" position. Not sure how I'd go about fixing that quite yet, though.

The motherboard might have a fault that results in the relay being permanently turned on. To see if that's the case, you will need to do some measurements. Refer to the diagram at [here], which shows what is expected to be observed when the motherboard is simply powered on. Are you able to make those measurements?

I'll expand on what I wrote:

If DIN pins 1 and 3 are closed circuit (zero Ohms) when the motherboard is unpowered, then the relay is at fault. In which case, try lightly tapping the relay with the handle of a screwdriver.

If DIN pins 1 and 3 are open circuit (infinite Ohms) when the motherboard is unpowered, and closed circuit (zero Ohms) when the motherboard is powered, then a motherboard fault has the relay turned on all the time. In which case, make the measurements that I indicated in my previous post in order to work out the source of the problem.